Project Abstract Speech communication impairments, including persistent difficulties in understanding and interpreting verbal information during conversation, are a hallmark of childhood autism. Speech-based communication unfolds over time, and speech comprehension relies on: (1) anticipation of incoming speech as a means of predicting its content and (2) temporal integration of speech so a listener can accumulate information over time to decode meaning in an extended utterance. Prominent theories of autism spectrum disorder (ASD) posit deficits in contextual and global information processing, which are germane to the anticipation and integration of information during communication. Late childhood is a crucial period for increased and more complex social interactions, including extended discourse between communication partners. Little is known regarding anticipatory and integrative components of speech processing, and their contribution to social communication (SC) deficits, in children with ASD. Advances in experimental design and computational analysis of human brain imaging data provide a unique opportunity to probe dynamic components of speech comprehension during naturalistic social interactions in children with ASD, which are difficult to ascertain using behavioral methods alone. Leveraging innovative fMRI experimental designs, we will for the first time investigate anticipatory and integrative aspects of naturalistic communication in children with ASD. Our overarching goal is to identify the neurocognitive mechanisms underlying speech comprehension deficits during naturalistic communication in children with ASD. The proposed studies include both speaker-listener brain coupling and temporal integration paradigms and build on our high-impact line of voice perception research in children with ASD. We hypothesize that children with ASD will show deficits in dynamic mechanisms of speech comprehension including anticipatory, reactive, and integrative processing with dissociable patterns of dysfunction in the default mode network (DMN), anchored in medial prefrontal cortex (mPFC) and posterior medial cortex (PMC), and lateral frontoparietal network (LFPN). While the DMN is often considered a ?task- negative? network, evidence shows that the DMN is crucial for processing social information, including narrative processing, and is closely linked to SC deficits in ASD. We hypothesize a link between these dynamic mechanisms and comprehension of global, but not local, narrative information, supporting the Weak Central Coherence model of ASD. Findings will provide new insights into speech comprehension impairments and advance our understanding of the role of the DMN in SC and ASD. Our studies will provide critical information regarding the neurobiological origins of communication impairments in ASD and will inform the development of age-appropriate treatment for older children with ASD. Our aims are in line with the NIH directive on Autism Research (PA-18-400), emphasizing brain mechanisms and sophisticated measures of social communication.